re: Flow integration
| Newsgroups | gmane.comp.lang.smalltalk.squeak.foundation |
|---|---|
| Message-ID | <[email protected]> |
Craig Latta <[email protected]> wrote: > > Hi Colin-- > > > This process would be appropriate for a monolithic image, but it > > doesn't fly very well in a modular universe. > > Normally I'd agree, but since virtually all the applications we're > talking about recently *were* in a monolithic image (we're still busy > taking them all out :), I think it's much less work right now to proceed No they were not. VH (a project of Bluefish) is not in the image and use the Sockets heavily. My point being that I agree fully with Colin, Andreas and Cees on this one. > the "monolithic" way. For other applications which were never part of > the release image, we can adapt just by making sure they're loaded, > installing Flow, and fixing what breaks. I effectively did that for > everything in 3.2final, and it was easy. First of all this would require we round up all apps out there that do depend on this - and note that there may be non open source apps out there too... Doesn't sound like a plan to me. Secondly this would mean adopting Flow before we actually have tried it out heavily. That also seems foolish. Instead I agree with the process Colin outlined with the little exception that I would like to make clear that the original networking package would be the one "blessed" from step 2 and forwards. First when Flow attracts enough users and has proven itself "in battle" we can consider moving the blessing over to Flow - if we decide that Flow is a better base package than the original. Note also that I also agree with Andreas/Colin that instead of renaming the old classes (thus breaking any code outside the image) why not simply having new names in Flow? It seems very obvious to me. regards, Göran